Urban Sprawl Prediction and Analysis for Atlanta
What is Urban Sprawl?
Urban sprawl, also called sprawl or suburban sprawl, is the rapid expansion of the geographic extent of cities and towns, often characterized by low-density residential housing, single-use zoning, and increased reliance on the private automobile for transportation.
The urban sprawl phenomenon, born in America in the early 1960s. From the 1950s until today, almost 8 million hectares agricultural land were lost.
According to a report by Smart Growth America, Atlanta is one of the most sprawling big metro regions in the country
The main negative effects of urban sprawl are, apart from the lack of planning for the city expansion, the high land use corresponding to a low population density.
Characteristics of Urban Sprawl
Study Area
Atlanta is the capital of the U.S. state of Georgia and it is the ninth largest metropolitan area in the country by population. Given that almost 60% of the state is urbanized and rapidly growing, for this study, the enumeration unit used was Census County Divisions (CCD). 197 out of 586 CCDs in the Northern part of the state were selected to study the urban sprawl. All the layers were clipped to the bounding region of the selected dissolved CCDS.
Study Area, Enumeration Unit: Census County Divisions (CCDs)
Data
The administrative boundaries and the Expressways road network data were downloaded from the U.S. Census Bureau and the Atlanta Regional Commission. The National Land Cover Database (NLCD) which has land cover and land cover change at a 30m resolution with a 16-class legend based on a modified Anderson Level II classification system was downloaded from the Multi-Resolution Land Characteristics (MRLC) consortium. These 16 classes were remapped into 5 classes. The 30m resolution DEM was downloaded from U.S Geological Survey to calculate the slope.
Methodology
Reclassification
After finalizing the study area, the land cover types were reclassified into five categories for prediction namely
- Forests
2. Agriculture/Barren Lands
3. Urbanized Areas
4. Wetlands
5. Water Bodies
2001 2006 2011 2016 2019
The QGIS Plugin: MOLUSCE
The Asia Air Survey released MOLUSCE (Modules for Land Use Change Evaluation) which is a plug-in for QGIS used extensively for land use/cover change analysis.
GUI Tabs
Area Change ANN Modelling Cellular Automata Simulation
The LULC data were collected for 5 years: 2001, 2006, 2011, 2016, and 2019. The initial and final years were used as inputs whereas the epochs in other years were used as spatial variables. First, the model computes land use/cover changes between two time periods. The built-in ANN Learning algorithm analyses the reached accuracy on training and validation sets of samples and stores the best neural network in its memory. The training process finishes when the best accuracy is reached. Then finally, a simulated (projected) land use/cover map is produced based on a Monte Carlo Cellular-automata modelling approach.
Methodology steps
Results
- Calculating Urban Density for CCDs
Urban Density
Initially, they were also reclassified into 2 classes namely as an urban and non-urban land-type. Using zonal statistics, the urban pixels were aggregated and the proportion of urban areas in each land type was calculated.
Urbam Density 2001 vs 2019
2. Direction of Growth
The potential direction of growth and the existing data depicted that the Northeastern part of Atlanta is booming with development.
3. Prediction of sprawl
2001 2019 2037
Simulated Urban Density for the year 2037
The results of the analysis for the simulated 2037 model show that the urban density increases along the major road networks which can be seen in the map above. This was significant given that the road network data was not given as an input as a spatial variable in the model. It was also observed that the direction of growth or the quadrant that had the most urban area was in the northeast from the centre of Atlanta and the simulated map predicts the same.
827.604 square kilometers of Forest deforestation!
1332.356 square kilometers of Urban development
Conclusion
In this project, the spatiotemporal urban development changes were analysed and projected for a scenario 15 years into the future based on based on past LULC data for the city of Atlanta and its surrounding CCDs. The potential direction of growth and the existing data depicted that the Northeastern part of Atlanta is booming with development. According to a recent study released by the U.S. Forest Service, Georgia is losing more trees than any other state in the nation. It was unfortunate to see the rapid degradation of forest areas of almost 828 squared kilometres in 18 years. The growing Urbanization is inevitable, hence is it important to have smart growth of cities instead of urban sprawl. Some great characteristics of smart growth in cities include having different housing options, accessible sidewalks, bike lanes and public transportation. Sustainable development and preserving natural resources is vital.
Future Scope
In future studies, biophysical and socio-economic data such as road network, rivers, topography, and even population could be used to predict the the landscape patterns. Additionally, the study could delve deeper by conducting the research on cities with different sizes and demographics. Further research could also study urban sprawl at a smaller resolutionto provide more acurate results for overcoming the negative consequences of this universal phenomenon.
References
1. The impacts of Atlanta’s urban sprawl on forest cover and fragmentation
2. An Analysis of Urban Sprawl and Prediction of Future Urban Town in Urban Area of Developing Nation: Case Study in India
3. The Problem of Urban Sprawl
4. A remote sensing and GIS-based analysis of urban sprawl in Soran District, Iraqi Kurdistan Problem of Urban Sprawl
5. Geographical Modeling of Spatial Interaction between Built-Up Land Sprawl and Cultivated Landscape Eco-Security under Urbanization Gradient Problem of Urban Sprawl
6. Evaluation of Urban growth and expansion using Remote sensing and GIS
7. Urban sprawl: metrics, dynamics and modelling using GIS
8. Mining GIS Data to Predict Urban Sprawl https://arxiv.org/ftp/arxiv/papers/2103/2103.11338.pdf
